HIV-1 associated dementia is an important complication of viral infection and a cause of significant morbidity and mortality. The underlying feature of HIV-1 induced neurological disease revolves around two processes: a) productive replication of the virus in macrophages in the brain, leading to encephalitis, and b) neuronal degeneration resulting from the action of secreted byproducts released from infected macrophages, leading to dementia. Our earlier studies on microarray analysis demonstrated a significant up-regulation of the chemokine, CXCL10 in the neurons in the brains of macaques with SHIV-encephalitis. Furthermore, we also showed that binding of viral glycoprotein (gp120) to the neurons led to the induction of CXCL10 and subsequent apoptosis of these cells (87). We hypothesize that elevated levels of CXCL10 in the CNS of HIV/SIY-infected subjects could contribute to the neuronal dysfunction observed in AIDS dementia through direct interactions with chemokine receptors expressed on neurons. This proposal is aimed at exploring an innovative and novel type of gene therapy involving delivery of antisense CXCL10 DNA targeted to neurons in the brain with a view to abrogating neuronal apoptosis. We will use the unique approach of feline immunodeficiency virus (FFV) based lentivirus vectors (2) to transduce the neurons adjacent to the white matter tract in the mouse brain. The long-term goal of this application (not a part of this proposal) is to develop therapeutic intervention strategy for HAD in SHTV-infected macaques. In this application we will test the hypothesis in two specific aims: 1) To examine the effect of CXCL10 over-expression on neuronal dysfunction in the brains of mice and, 2) To abrogate neuronal apoptosis by introduction of antisense CXCL10 DNA in mouse brain.